Originally published May 7.
By Turna Ray
Whenever anyone questions why genomically guided personalized medicine hasn't yet delivered on its promise to make healthcare cheaper and more efficient, the blame often falls on the physicians. Anachronistic doctors' unwillingness to adopt medical advancements, their discomfort with genomic data, and their lack of genomic education are commonly cited as the main reasons that pharmacogenomics hasn't taken off.
An oft-cited survey by Medco and the American Medical Association of nearly 11,000 physicians found that although 98 percent are aware that patient genetic profiles may influence therapy, only 10 percent believe they are "adequately informed" about the use of genetic information in practice.
According to one physician and early-adopter in the personalized medicine space, his colleagues just need to be reminded why they got into medicine in the first place, and the intellectual challenge of genomic medicine itself can reignite that spark. Earlier this year, Eric Topol, the director of the Scripps Translational Science Institute, received a grant from Life Technologies Foundation to create an online educational program to credential doctors in genomics.
The effort, called the Association for Genomic Medicine, is currently working within its accreditation board, comprised of key genomics leaders, to develop a curriculum, and is slated to launch by the end of the year (PGx Reporter 03/10/10).
Topol, who implemented PGx-guided dosing strategies for the anti-platelet drug Plavix within Scripps Health last year, discussed with Pharmacogenomics Reporter this week about where things stand with the so-called "Genomic Medicine Academy," and the broader challenges associated with getting doctors up to speed in the field of personalized medicine.
Below is an edited transcript of the interview.
What are you trying to achieve through this certification course?
The aim is to try to credential as many physicians as possible in genomic medicine. We're working on a curriculum, and we have a board that we've formed of international experts in genomics and medicine. And we're going to have a remote learning instrument through the web, what we internally call 'Genomic Medicine University,' or 'Genomic Medicine Academy.' We'll have a couple of different levels of genomic teaching in a really exciting and dynamic web-based way, with testing and authentication to make sure that it's really the physician that's going through the process.
How much grant money did you get from Life Technologies for this?
The $300,000 seed grant is enough to get us started with the curriculum development, assembly of the board, our mission, and our goals. It isn't necessarily enough to finish everything, but it will get the process underway.
In surveying the state of current physician education, what deficiencies did you identify in terms of genomics learning?
Well [the deficiencies] are profound and they are global. We have a course, called 'Future of Genomic Medicine,' where we have 400 people come every year, and this is our third round in March. We did a survey of everybody there about what the unmet needs in education are and the answer to that question was that they were profound and almost everyone thought that a web-based program would work. The minimum that people thought physicians would devote to this was at least eight hours. And many thought [doctors would devote] more than 40 hours, which is impressive.
There was also a big survey that was published by the American Medical Association, in conjunction with Medco, and that survey found that 90 percent of physicians said they felt completely uncomfortable using any genomic data in treating their patients. That's not a good thing.
At the same time, that survey showed that the majority of physicians were interested in using pharmacogenomics in their practices.
What we're learning, though, in the more developed [PGx] areas, is that even when you give them the data, they feel uncomfortable. They need to get up to speed. The other thing is even young physicians, those who graduated medical school in the last five years, they are not so different than the older ones.
Currently, there is no entity that's really going after this education mission, which is a big one. And we're very fortunate that Life Technologies is getting behind this, and hopefully, we'll get other companies and foundations to help support this. We can only go so far.
Since this is funded by Life Technologies, will your curriculum include information on whole genome sequencing and next-generation sequencing?
Oh, yes. Only one part of the curriculum is about pharmacogenomics. A lot of it is about the genome and about the tools, such as sequencing, high-throughput genotyping, methylation, and all the different things that are done to unravel genomics, proteins, and metabolites.
As more groups like Medco invest in the area of personalized medicine, by launching e-health systems that alert the physicians when their patients need a pharmacogenomic test, and deploy pharmacists to discuss with doctors their patients' genotypes, some people think this is putting the doctor in a box. Some feel that doctors may be resistant to this kind of pressure. In your interactions with other doctors have you seen this?
Physicians' number one concern is not [about] getting up to speed in genomics. Their more practical concerns are reimbursement issues. We have to override that and have something that's really exciting and gets them back to why they went back into medicine in the first place, which is the science and the intellectual stimulation and that when they are empowered with this information they are going to do well for people.
The other concern [for doctors] is that the consumers are getting more quickly oriented in genomics than their physicians, by virtue of a lot of people having genotyping and genome-wide scanning and that sort of thing. So, they are showing up to their doctor saying, 'I can't take this medicine because I have such and such genotype.' They are catching physicians off guard, who are not up to speed. I think that's helping a little bit, actually.
Diagnostics companies are increasingly conducting direct-to-consumer marketing. How have doctors reacted to DTC marketing that brings more people to their offices asking about genetic tests, such as testing for BRCA mutations?
It's very different by discipline. Oncologists who take care of breast cancer patients are very comfortable with [DTC marketing of genetic tests.] But that's a very small proportion of physicians in the country, around 800,000 physicians. There are a few examples like [BRCA] but they are few and far between.
Recently in a paper published in The Lancet, researchers analyzed [Stanford University Professor of Bioengineering] Stephen Quake's sequenced genome for clinically meaningful disease predisposition and pharmacogenomics data. In speaking with Russ Altman, [a professor of bioengineering, genetics, and medicine at Stanford University and] one of the authors of that paper, he noted that there was more uncertainty in the disease predisposition analysis than in the PGx data. In designing this curriculum, how will you handle the uncertainty of characterizing disease predisposition?
As you know from the Quake genome, it kind of broke down the three areas. One was genome-wide association study-like common variants, although they had a bigger sweep of 2.5 million SNPs. The other were rare variants, which up until now we really haven’t had a window into. And then there was the pharmacogenomic story, where there were 63 interesting drug interactions that were noted. That's a pretty good categorization.
The whole common variants story was much more suspect than what was originally thought. So, the things that are panning out are rare variants and pharmacogenetic variants. As we go forward, trying to get the medical community up to speed, that's a pretty reasonable way to try to go over the data, the way it was presented with Steve's data … In passing I should mention that Russ is on the board of the Association of Genomic Medicine [created with the Life Technologies grant]. He's terrific; he's the guru of pharmacogenetics.
In terms of the disease risk data, specifically, it seems like doctors will need support from genetic counselors and nurses in counseling and explain results to patients. Will the course you are developing be applicable to other medical professionals, other than doctors?
It will be directed to physicians, but obviously if nurses and other medical professionals want to do it, then they are more than welcome. It's hard to have it at many different levels. So, we want to first get to physicians, but it will be open to anyone who wants to get into the curriculum.
Do you think that with the genomic data, doctors can actually change patient behavior? Some think that it won't. For example, despite a family history and genetic variants suggesting heightened risk for vascular disease, Dr. Quake has chosen to hold off on going on statins despite his doctors' recommendation. Why should doctors invest in learning this if ultimately it's not going to change how their patients behave?
The motive is that it's science and it's intellectually stimulating. We need more of that for physicians because they are more challenged by dealing with insurance companies and having their reimbursement gutted, and that sort of thing. It's also a means to provide better care, state-of-the art care. The ones who are up to speed, are going to be viewed as physicians that are keeping up with the field. Also, how are you going to take care of patients who are already armed with this information, and are you going to look like you're not prepared when patients come in [discussing their genetic data] and you don't even know what they are talking about?
Now, we have a different model. There is SNPedia, there are genome scans, there are [going to be] iPhones you can load up all your genome data on. It's a different world, and we never had that in medicine before.
Is there anything new to report from your efforts at Scripps Health to further the adoption of pharmacogenomics among doctors?
The program that we started last September was the Plavix genotyping, and that's been really a great success. As you know, in March, the US Food and Drug Administration put a black box warning about [genetic response] on Plavix's label. So, that model, of finding where the data are compelling and checking systematically in patients, we're going to do that in other areas in the months ahead … We're gearing up to do a couple more later this year, but I can't tell you what they are yet.
Do the doctors who are now genetically testing patients in the context of administering Plavix feel comfortable doing this? And how are the patients handling it?
I think patients are really receptive, I think they like the fact their therapy is getting individualized. And the physicians are getting really used to it now. They are really very supportive. They weren't initially, but they are really coming around to how important this is. It's great to see that.
What were their concerns initially?
Oh, [they worried] they wouldn't be able to interpret the data, they wouldn't know what to do with it, and would it really make a difference? Is it going to make life more complicated? The test results come a few days after they talk with the patient, and they have to call and change the prescription. It's hassle stuff [they're worried about]. But with education, it's been smoother than expected, and it's starting to become routine … It's a good model. It's a highly actionable thing. And there is this intermediate test, platelet function testing, and that can be used. So, there are a lot of different ways to deal with the data, and each physician really feels comfortable doing this now. It's a really great step forward.